You are on page 1of 3

Implicit functions

Based on lecture notes by James McKernan


Consider the curve y 2 = x in the plane R2 ,
C = { (x, y) R2 | y 2 = x }.
This is not the graph of a function, and yet it is quite close to the graph of a
function.
Given any point on the graph, lets say the point (2, 4), we can always find open
intervals U containing 2 and V containing 4 and a smooth function f : U V
such that C (U V ) is the graph of f .

Indeed, take U = (0, ), V = (0, ) and f (x) = x. In fact, we can do this


for any point on the graph, apart from the origin. If it is above the x-axis, the
function above works. If the
in is below the x-axis, replace
point we are interested

V by (0, ) and f (x) = x, by g(x) = x.


How can we tell that the origin is a point where we cannot define an implicit
function? Well away from the origin, the tangent line is not vertical but at the
origin the tangent line is vertical. In other words, if we consider
F : R2 R,
given by F (x, y) = y 2 x, so that C is the set of points where F is zero, then
DF (x, y) = (1, 2y).
The locus where we run into trouble, is where 2y = 0. Somewhat amazingly this
works in general:
Theorem 1 (Implicit Function Theorem). Let F : Rn+m Rm be a C 1 -function.
Suppose that
(~a, ~b) S = { (~x, ~y ) Rn+m | F (~x, ~y ) = ~0 }.
Assume that

det

Fi
yj


6= 0.

Then we may find open subsets ~a U Rn and ~b V Rm , where U V Rn+m


and a function f : U V such that S (U V ) is the graph of f , that is,
F (~x, ~y ) = ~0

if and only if

~y = f (~x).

where ~x U and ~y V .
Lets look at an example. Let
F : R3 R,
be the function
F (x1 , x2 , y) = x31 x2 x2 y 2 + y 5 + 1.
Let
S = { (x1 , x2 , y) R3 | F (x1 , x2 , y) = 0 }.
1

Then (1, 3, 1) S. Lets compute the partial derivatives of F ,




F
2
(1, 3, 1) = 3x1 x2
=9
x1
(1,3,1)


F
3
2
=0
(1, 3, 1) = (x1 y )
x2
(1,3,1)


F
= 11.
(1, 3, 1) = (2x2 y + 5y 4 )
y
(1,3,1)

So
DF (1, 3, 1) = (9, 0, 11).
Now what is important is that the last entry is non-zero (so that the 1 1 matrix
(1) is invertible). It follows that we may find open subsets (1, 3) U R2 and
1 V R and a C 1 function f : U V such that
F (x1 , x2 , f (x1 , x2 )) = 0.
It is not possible to write down an explicit formula for f , but we can calculate the
partial derivatives of f .
Define a function
G : U R,
by the rule
G(x1 , x2 ) = F (x1 , x2 , f (x1 , x2 )) = 0.
On the one hand,
G
=0
and
x1
On the other hand, by the chain rule,

G
= 0.
x2

F x1
F x2
F f
G
=
+
+
x1
x1 x1
x2 x1
x3 x1
Now
x1
=1
x1

x2
= 0.
x1

and

So
F

f
1
= x
.
F
x1
x
3

Similarly
F

f
2
= x
.
F
x2
x
3

So
F
(1, 3, 1)
f
9
1
(1, 3) = x
= ,
F
x1
11
x (1, 3, 1)
3

and
F
(1, 3, 1)
f
0
2
(1, 3) = x
=
= 0.
F
x2
11
x (1, 3, 1)
3

Blackboard 2. Let A Rn be an open subset and let f : Rn R be a function.


The directional derivative of f in the direction of the unit vector u
is
f (p + h
u) f (p)
Du f (p) = lim
.
h0
h
If u
= ei then,
f
(p),
Dei f (p) =
xi
the usual partial derivative.
Proposition 3. If f is differentiable at p then
Du f (p) = Df (p) u
.
Proof. Since A is open, we may find > 0 such that the parametrised line
r : (, ) A,
given by r(h) = f (p) + h
u is entirely contained in A. Consider the composition of
r and f ,
f r : R R.
Then


d(f r)
Du f (p) =
(0)
dh
= D(r(0)) Dr(0)
= Df (p) u
.

Note that we can also write


Du f (p) = f (p) u
.
Note that the directional derivative is largest when
f (p)
,
u
=
kf (p)k
so that the gradient always points in the direction of maximal change (and in
fact the magnitude of the gradient, gives the maximum change). Note also that
the directional derivative is zero if u
is orthogonal to the gradient and that the
directional derivative is smallest when
f (p)
u
=
.
kf (p)k
Proposition 4. If f (p) 6= 0 then the tangent hyperplane to the hypersurface
S = { q Rn | f (q) f (p) = 0 },
is the set of all points q which satisfy the equation

f (p)
pq = 0.